CN104865313A - Method and device for detecting glass breaking based on sound spectrum stripes - Google Patents
Method and device for detecting glass breaking based on sound spectrum stripes Download PDFInfo
- Publication number
- CN104865313A CN104865313A CN201510237822.6A CN201510237822A CN104865313A CN 104865313 A CN104865313 A CN 104865313A CN 201510237822 A CN201510237822 A CN 201510237822A CN 104865313 A CN104865313 A CN 104865313A
- Authority
- CN
- China
- Prior art keywords
- glass breaking
- audio data
- data sample
- unit
- striped
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention relates to the technical field of sound spectrum stripes, and in particular relates to a method and a device for detecting glass breaking based on sound spectrum stripes, which can collect audio data in a certain space without installing the detecting device on a glass product, thus being more flexible and convenient to use. The glass breaking is detected by making statistics on the number of the sound spectrum stripes; compared with the existing method for detecting glass breaking through inducing high frequency energy by piezoelectric ceramics, the method provided by the invention takes account of high frequency information and is also added with stripe detection, thus being higher in accuracy, omitting the necessary of specific hardware material such as a piezoelectric sensor, and being conductive to popularization for use.
Description
Technical field
The present invention relates to sound spectrum fringe technique field, particularly relate to a kind of detection method based on sound spectrum bar detection glass breaking and device.
Background technology
From bank, jeweler's shop to the house of household, the glassware such as windowpane, glass cabinet is ubiquitous, very important safety guarantee product in people's daily life, a lot of thief steals, so glass breaking alarm, glass break detection device are more and more important at safety-security area by destroying glass now.
In recent years, glass break detection device emerges in an endless stream, and is roughly divided into two large classes: acoustic control and oscillating mode are namely respectively to detect glass breaking sound characteristic and the stressed change of glass breaking to detect glass break event.The vibration that the many utilizations of oscillating mode specific hardware material produces when carrying out perception glass breaking, need be arranged on glassware, install and usable range all limited.The sound control type then main sound characteristic according to glass breaking realizes, and it can be arranged on each position of distance glassware certain distance, uses more flexible.
The high-frequency sound that the many employings of existing sound control type glass break detection device produce when detecting glass breaking occurs to detect glass breaking, if publication date is 2013-01-16, publication number is a kind of method that the Chinese invention of 102874212A discloses inspection vehicle glass breaking, it produces the high-frequency sound of 10k ~ 15k when utilizing piezoelectric ceramic piece to carry out perception glass breaking, judge that glass event occurs with this.The method only relies on and detects 10k ~ 15k high frequency because of usually judging glass break event, easily knows situation because other high-frequency sounds in daily produce by mistake, and needs specific piezoelectric ceramic device to carry out the generation of perception high-frequency sound, be unfavorable for promoting the use of.
Summary of the invention
Technical matters to be solved by this invention is: provide a kind of detection method based on sound spectrum bar detection glass breaking and device, improves the accuracy that glass breaking detects.
In order to solve the problems of the technologies described above, the technical solution used in the present invention is:
Based on a detection method for sound spectrum bar detection glass breaking, comprising:
Obtain voice data to be detected;
Described voice data is converted to corresponding segments sonogram data;
Add up the striped quantity of the predeterminated frequency scope in described segments sonogram data;
Judge that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
Based on a pick-up unit for sound spectrum bar detection glass breaking, comprising: the first acquisition module, modular converter, statistical module and judge module;
Described first acquisition module, for obtaining voice data to be detected;
Described modular converter, for converting corresponding segments sonogram data to by described voice data;
Described statistical module, for adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data;
Described judge module, for judging that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
Beneficial effect of the present invention is: the detection method based on sound spectrum bar detection glass breaking provided by the invention and device can gather voice data in certain space, without the need to being arranged on glassware by pick-up unit, uses more flexible.Glass breaking is detected by adopting statistics sound spectrum striped number, compare existing employing piezoelectric ceramics and carry out induction of high frequency energy to detect the method for glass breaking, not only take into account high-frequency information, and with the addition of bar detection, accuracy is higher, also eliminate the necessity using the specific hardware materials such as piezoelectric sensor, be more conducive to promote the use of.
Accompanying drawing explanation
Fig. 1 is a kind of detection method flow chart of steps based on sound spectrum bar detection glass breaking of the specific embodiment of the invention;
Fig. 2 is sonograph and the profile switching process schematic diagram of the specific embodiment of the invention;
Fig. 3 is that the sonograph of the specific embodiment of the invention converts two-dimensional array schematic diagram to;
Fig. 4 is that the two-dimensional array of the specific embodiment of the invention converts one-dimension array schematic diagram to;
Fig. 5 is that the one-dimension array of the specific embodiment of the invention converts curve synoptic diagram to;
Fig. 6 is a kind of structure of the detecting device schematic diagram based on sound spectrum bar detection glass breaking of the specific embodiment of the invention;
Label declaration:
10, the first acquisition module; 20, modular converter; 30, statistical module; 40, judge module.
Embodiment
By describing technology contents of the present invention in detail, realized object and effect, accompanying drawing is coordinated to be explained below in conjunction with embodiment.
The design of most critical of the present invention is: the striped quantity of adding up the predeterminated frequency scope in described segments sonogram data; Judge that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
Please refer to Fig. 1, is a kind of detection method flow chart of steps based on sound spectrum bar detection glass breaking of the specific embodiment of the invention, specific as follows:
Based on a detection method for sound spectrum bar detection glass breaking, comprising:
Obtain voice data to be detected;
Described voice data is converted to corresponding segments sonogram data;
Add up the striped quantity of the predeterminated frequency scope in described segments sonogram data;
Judge that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
From foregoing description, beneficial effect of the present invention is: the detection method based on sound spectrum bar detection glass breaking provided by the invention can gather voice data in certain space, without the need to being arranged on glassware by pick-up unit, uses more flexible.Glass breaking is detected by adopting statistics sound spectrum striped number, compare existing employing piezoelectric ceramics and carry out induction of high frequency energy to detect the method for glass breaking, not only take into account high-frequency information, and with the addition of bar detection, accuracy is higher, also eliminate the necessity using the specific hardware materials such as piezoelectric sensor, be more conducive to promote the use of.
Further, described " obtaining voice data to be detected " is specially: the voice data to be detected obtaining preset audio length;
The acquisition methods of the value of described preset audio length is:
Obtain the audio data sample of multiple glass breaking;
End-point detection is carried out to described audio data sample;
Draw the probability distribution graph of the audio data sample after end-point detection;
Obtain the value of preset audio length corresponding to predetermined probabilities value range in described probability distribution graph.
Seen from the above description, by the audio data sample of the multiple glass breaking of above-mentioned acquisition, join probability distribution plan can draw the audio frequency length (being the value of preset audio length) of general glass breaking sound, in actual experiment, gather the audio data sample quantity of 1000 glass breakings, according to said method, show that audio frequency length meets the requirement of probability more than 90% in 0.6s-0.8s scope, wherein optimal value is 0.7s.Described audio frequency length is time span.The acquisition of the value of described preset audio length only obtains in first time is detected, if carry out repeated detection, then the value of the preset audio length directly using first time to obtain, without the need to again obtaining.
Further, the acquisition methods of " the first pre-set threshold value " in " predeterminated frequency scope " in described " adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data " and described " judging that whether described striped quantity is more than the first pre-set threshold value " is:
Gather the audio data sample of multiple glass breaking and the audio data sample of everyday sound;
The audio data sample obtaining described glass breaking is at the striped number of different frequency scope and the audio data sample of the described everyday sound striped number in different frequency scope;
The audio data sample of drawing described glass breaking is in the fringe number object probability distribution graph of different frequency scope and the audio data sample of the described everyday sound fringe number object probability distribution graph in different frequency scope, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value.Wherein, misclassification rate refers to that everyday sound is identified as the percent of glass breaking sound, and False Rate refers to the percent that glass breaking sound is not identified.
Seen from the above description, technical scheme of the present invention considers that the voice data of glass breaking may be close with everyday sound, easily cause erroneous judgement or know by mistake, therefore technical scheme of the present invention adopts and gathers the audio data sample of multiple glass breaking and the audio data sample of everyday sound, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value, guarantee the predeterminated frequency scope that draws and the first pre-set threshold value more accurate.In actual experiment, the quantity of the audio data sample of glass breaking and the audio data sample of everyday sound is 1000, and the predeterminated frequency scope drawn is 3445Hz-17226Hz, and the first pre-set threshold value is 14.The acquisition of described " predeterminated frequency scope " and " the first pre-set threshold value " only obtains in first time is detected, if carry out repeated detection, then directly uses " predeterminated frequency scope " and " the first pre-set threshold value " that first time obtains, without the need to again obtaining.
Wherein, judge the method for striped as:
By section audio data temporally length x (this time span is the time span reference value of a stripe, probably add up according to great amount of samples, this time span of testing a stripe is: 0.1s) be divided into n segment, obtain the striped number of each segment respectively, striped number and the striped number being this segment data of n segment.
Length x statistical method is as follows:
1, observe the approximate size of striped in sonograph, formulate substantial distance term of reference, between 0.05s-0.2s;
2, a stripe time span term of reference is carried out being divided into different length grade, such as 0.05s, 0.1s, 0.15s, 0.2s;
3, according to the striped number in striped different length grade statistics glass sample and the striped number probability distribution of daily life sample;
4, False Rate and the misclassification rate of different length grade is assessed; As shown in table 1 below, when the False Rate of optimum is 5.2% and misclassification rate is 9.8%, striped time span is 0.1s.
Table 1
5, the optimum False Rate length scale judgment criteria corresponding with misclassification rate is obtained.Be specially: predeterminated frequency scope is 3445Hz-17226Hz, striped time span scope is 0.05s-0.2s, is preferably 0.1s.
Further, the detection method of described " adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data " is specially:
Convert described segments sonogram data to one-dimensional curve;
Add up transverse width on described one-dimensional curve and be less than or equal to a stripe breadth extreme, be longitudinally highly greater than the spike number of the second pre-set threshold value, described spike number is striped quantity.
Further, described " converting described segments sonogram data to one-dimensional curve " is specially:
Take transverse axis as the time, the longitudinal axis converts described segments sonogram data to two-dimensional array for frequency;
The data of same frequency in described two-dimensional array are added up, obtains one-dimension array;
Data rear in described one-dimension array and last data are subtracted each other between two, obtains new one-dimension array;
Described new one-dimension array converts one-dimensional curve to.
Wherein, as shown in Figure 2, transfer process is specific as follows:
Be that the sound spectrum data plot A of x transforms into simple one-dimensional curve figure B by length.
In detection figure B curve, the number of spike is striped number.
Conversion process and principle as follows:
In sonograph, what transverse axis represented is time scale, and what the longitudinal axis represented is frequency scale, and color value represents energy size.Bar detection mainly detects based on the horizontal stripe longitudinally formed because of energy variation in sonograph, the change curve being longitudinal energy in sonograph A represented in upper figure B, and its acquisition methods is as follows:
1) as Fig. 3, sonograph information adopts the mode of image display for shown in sonograph A, it can represent with a two-dimensional array by the mode that data store, i.e. two-dimensional array A ＇, represent time scale with the horizontal ordinate of array, ordinate represents frequency scale, and the numerical value inside array represents energy value, namely the color value in sonograph, namely the change of numerical values recited represents the change of each color value.
2) as Fig. 4-5, conversion process is as follows:
Step 1: the data accumulation of same frequency (namely ordinate is identical) in two-dimensional array A ＇ is become a numerical value, obtains one-dimension array B ＇.
Step 2: by one-dimension array below data and above data subtract each other between two (first is 0), obtain the new one-dimension array B (namely array index represents frequency scale) representing difference and change.
It (is laterally array index that step 3: array B can be expressed as curve B, be longitudinally numerical values recited), what it reacted is that in the unit interval, acoustic energy, along with the change of frequency, if form spike, then represents energy from low to high, arrive again low, namely be expressed as longitudinally from dark to bright on sound spectrum, again from bright to secretly, thus form striped, the shape of spike also reacts the information such as width, the depth of striped simultaneously, therefore can arrange spike testing conditions to arrange the Testing index of striped.In the present invention, spike testing conditions is: transverse width is satisfied is less than or equal to a stripe breadth extreme, a described stripe breadth extreme is that 11 (in the present invention, sonograph frequency range is 0-22050KHz, be translated between 0-256 scale area, 11 are about the interval width corresponding to 1KHz), be longitudinally highly greater than 4 the difference of energy value in sonograph (4 be).The spike number of detection curve B can react the striped number in sonograph A.
Please refer to Fig. 6, is a kind of structure of the detecting device schematic diagram based on sound spectrum bar detection glass breaking of the specific embodiment of the invention, specific as follows:
Based on a pick-up unit for sound spectrum bar detection glass breaking, comprising: the first acquisition module 10, modular converter 20, statistical module 30 and judge module 40;
Described first acquisition module 10, for obtaining voice data to be detected;
Described modular converter 20, for converting corresponding segments sonogram data to by described voice data;
Described statistical module 30, for adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data;
Described judge module 40, for judging that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
From foregoing description, beneficial effect of the present invention is: the pick-up unit based on sound spectrum bar detection glass breaking provided by the invention can gather voice data in certain space, without the need to being arranged on glassware by pick-up unit, uses more flexible.Glass breaking is detected by adopting statistics sound spectrum striped number, compare existing employing piezoelectric ceramics and carry out induction of high frequency energy to detect the method for glass breaking, not only take into account high-frequency information, and with the addition of bar detection, accuracy is higher, also eliminate the necessity using the specific hardware materials such as piezoelectric sensor, be more conducive to promote the use of.
Further, described first acquisition module comprises the first acquiring unit, end-point detection unit, drawing unit and second acquisition unit;
Described first acquiring unit, for obtaining the audio data sample of multiple glass breaking;
Described end-point detection unit, for carrying out end-point detection to described audio data sample;
Described drawing unit, for drawing the probability distribution graph of the audio data sample after end-point detection;
Described second acquisition unit, for obtaining the value of the preset length that predetermined probabilities value range is corresponding in described probability distribution graph.
Seen from the above description, by the audio data sample of the multiple glass breaking of above-mentioned acquisition, join probability distribution plan can draw the audio frequency length (being the value of preset length) of general glass breaking sound, in actual experiment, gather the audio data sample quantity of 1000 glass breakings, according to said method, show that audio frequency length meets the requirement of probability more than 90% in 0.6s-0.8s scope, wherein optimal value is 0.7s.Described audio frequency length is time span.
Further, the pick-up unit that the present invention is based on sound spectrum bar detection glass breaking described in also comprises acquisition module, the second acquisition module and drafting module;
Described acquisition module, for the audio data sample of the audio data sample and everyday sound that gather multiple glass breaking;
Described second acquisition module, for the audio data sample that obtains described glass breaking at the striped number of different frequency scope and the described everyday sound striped number in the audio data sample of different frequency scope;
Described drafting module, for the audio data sample of drawing described glass breaking in the fringe number object probability distribution graph of different frequency scope and the audio data sample of the described everyday sound fringe number object probability distribution graph in different frequency scope, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value.
Seen from the above description, technical scheme of the present invention considers that the voice data of glass breaking may be close with everyday sound, easily cause erroneous judgement or know by mistake, therefore technical scheme of the present invention adopts and gathers the audio data sample of multiple glass breaking and the audio data sample of everyday sound, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value, guarantee the predeterminated frequency scope that draws and the first pre-set threshold value more accurate.The predeterminated frequency scope drawn in actual experiment is 3445Hz-17226Hz, and the first pre-set threshold value is 14.
Further, described statistical module comprises converting unit and statistic unit;
Described converting unit, for converting described segments sonogram data to one-dimensional curve;
Described statistic unit, is less than or equal to a stripe breadth extreme for adding up transverse width on described one-dimensional curve, and be longitudinally highly greater than the spike number of the second pre-set threshold value, described spike number is striped quantity.
Further, described converting unit, specifically for take transverse axis as the time, the longitudinal axis converts described segments sonogram data to two-dimensional array for frequency; The data of same frequency in described two-dimensional array are added up, obtains one-dimension array; Data rear in described one-dimension array and last data are subtracted each other between two, obtains new one-dimension array; Described new one-dimension array converts one-dimensional curve to.
Embodiments of the invention one are:
According to the above-mentioned method provided, gather the audio data sample quantity and 1000 and everyday sound audio data sample quantity of 1000 glass breakings, can obtain: predeterminated frequency scope is 3445Hz-17226Hz, first pre-set threshold value is 14, preset audio length meets the demands in 0.6s-0.8s scope, and wherein optimal value is 0.7s.Strip length is 0.1s.
Obtain the voice data to be detected of 0.7s time span;
Described voice data is converted to corresponding segments sonogram data;
The predeterminated frequency scope of adding up in described segments sonogram data is 3445Hz-17226Hz, and strip length is the quantity of 0.1s;
Judge that whether described striped quantity is more than 14, if so, then glass break event occurs.
In addition, the acquisition of described " preset audio length ", " predeterminated frequency scope " and " the first pre-set threshold value " only obtains in first time is detected, if carry out repeated detection, then directly use " preset audio length ", " predeterminated frequency scope " and " the first pre-set threshold value " that first time obtains, without the need to again obtaining.
In sum, a kind of detection method based on sound spectrum bar detection glass breaking provided by the invention and device, can gather voice data in certain space, without the need to being arranged on glassware by pick-up unit, uses more flexible.Glass breaking is detected by adopting statistics sound spectrum striped number, compare existing employing piezoelectric ceramics and carry out induction of high frequency energy to detect the method for glass breaking, not only take into account high-frequency information, and with the addition of bar detection, accuracy is higher, also eliminate the necessity using the specific hardware materials such as piezoelectric sensor, be more conducive to promote the use of.Consider that the voice data of glass breaking may be close with everyday sound, easily cause erroneous judgement or know by mistake, therefore technical scheme of the present invention adopts and gathers the audio data sample of multiple glass breaking and the audio data sample of everyday sound, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value, guarantee the predeterminated frequency scope that draws and the first pre-set threshold value more accurate.
The foregoing is only embodiments of the invention; not thereby the scope of the claims of the present invention is limited; every equivalents utilizing instructions of the present invention and accompanying drawing content to do, or be directly or indirectly used in relevant technical field, be all in like manner included in scope of patent protection of the present invention.
Claims (10)
1. based on a detection method for sound spectrum bar detection glass breaking, it is characterized in that, comprising:
Obtain voice data to be detected;
Described voice data is converted to corresponding segments sonogram data;
Add up the striped quantity of the predeterminated frequency scope in described segments sonogram data;
Judge that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
2. the detection method based on sound spectrum bar detection glass breaking according to claim 1, is characterized in that, described " obtaining voice data to be detected " is specially: the voice data to be detected obtaining preset length;
The acquisition methods of the value of described preset length is:
Obtain the audio data sample of multiple glass breaking;
End-point detection is carried out to described audio data sample;
Draw the probability distribution graph of the audio data sample after end-point detection;
Obtain the value of the preset length that predetermined probabilities value range is corresponding in described probability distribution graph.
3. the detection method based on sound spectrum bar detection glass breaking according to claim 1, it is characterized in that, the acquisition methods of " the first pre-set threshold value " in " predeterminated frequency scope " in described " adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data " and described " judging that whether described striped quantity is more than the first pre-set threshold value " is:
Gather the audio data sample of multiple glass breaking and the audio data sample of everyday sound;
Obtain the striped number of the striped number of the audio data sample of described glass breaking and the audio data sample of described everyday sound;
Draw the fringe number object probability distribution graph of the fringe number object probability distribution graph of the audio data sample of described glass breaking and the audio data sample of described everyday sound, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value.
4. the detection method based on sound spectrum bar detection glass breaking according to claim 1, is characterized in that, described " adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data " is specially:
Convert described segments sonogram data to one-dimensional curve;
Add up transverse width on described one-dimensional curve and be less than or equal to a stripe breadth extreme, be longitudinally highly greater than the spike number of the second pre-set threshold value, described spike number is striped quantity.
5. the detection method based on sound spectrum bar detection glass breaking according to claim 4, is characterized in that, described " converting described segments sonogram data to one-dimensional curve " is specially:
Take transverse axis as the time, the longitudinal axis converts described segments sonogram data to two-dimensional array for frequency;
The data of same frequency in described two-dimensional array are added up, obtains one-dimension array;
Data rear in described one-dimension array and last data are subtracted each other between two, obtains new one-dimension array;
Described new one-dimension array converts one-dimensional curve to.
6. based on a pick-up unit for sound spectrum bar detection glass breaking, it is characterized in that, comprising: the first acquisition module, modular converter, statistical module and judge module;
Described first acquisition module, for obtaining voice data to be detected;
Described modular converter, for converting corresponding segments sonogram data to by described voice data;
Described statistical module, for adding up the striped quantity of the predeterminated frequency scope in described segments sonogram data;
Described judge module, for judging that whether described striped quantity is more than the first pre-set threshold value, if so, then glass break event occurs.
7. the pick-up unit based on sound spectrum bar detection glass breaking according to claim 6, is characterized in that, described first acquisition module comprises the first acquiring unit, end-point detection unit, drawing unit and second acquisition unit;
Described first acquiring unit, for obtaining the audio data sample of multiple glass breaking;
Described end-point detection unit, for carrying out end-point detection to described audio data sample;
Described drawing unit, for drawing the probability distribution graph of the audio data sample after end-point detection;
Described second acquisition unit, for obtaining the value of the preset length that predetermined probabilities value range is corresponding in described probability distribution graph.
8. the pick-up unit based on sound spectrum bar detection glass breaking according to claim 6, is characterized in that, also comprises acquisition module, the second acquisition module and drafting module;
Described acquisition module, for the audio data sample of the audio data sample and everyday sound that gather multiple glass breaking;
Described second acquisition module, for the striped number of the audio data sample of the striped number and described everyday sound that obtain the audio data sample of described glass breaking;
Described drafting module, for the fringe number object probability distribution graph of the audio data sample of the fringe number object probability distribution graph and described everyday sound of drawing the audio data sample of described glass breaking, according to optimum False Rate and misclassification rate, obtain predeterminated frequency scope and the first pre-set threshold value.
9. the pick-up unit based on sound spectrum bar detection glass breaking according to claim 6, is characterized in that, described statistical module comprises converting unit and statistic unit;
Described converting unit, for converting described segments sonogram data to one-dimensional curve;
Described statistic unit, is less than or equal to a stripe breadth extreme for adding up transverse width on described one-dimensional curve, and be longitudinally highly greater than the spike number of the second pre-set threshold value, described spike number is striped quantity.
10. the pick-up unit based on sound spectrum bar detection glass breaking according to claim 9, is characterized in that, described converting unit, specifically for take transverse axis as the time, the longitudinal axis converts described segments sonogram data to two-dimensional array for frequency; The data of same frequency in described two-dimensional array are added up, obtains one-dimension array; Data rear in described one-dimension array and last data are subtracted each other between two, obtains new one-dimension array; Described new one-dimension array converts one-dimensional curve to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510237822.6A CN104865313B (en) | 2015-05-12 | 2015-05-12 | A kind of detection method and device based on sound spectrum bar detection glass breaking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510237822.6A CN104865313B (en) | 2015-05-12 | 2015-05-12 | A kind of detection method and device based on sound spectrum bar detection glass breaking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104865313A true CN104865313A (en) | 2015-08-26 |
| CN104865313B CN104865313B (en) | 2017-11-17 |
Family
ID=53911292
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510237822.6A Active CN104865313B (en) | 2015-05-12 | 2015-05-12 | A kind of detection method and device based on sound spectrum bar detection glass breaking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104865313B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108098586A (en) * | 2017-12-25 | 2018-06-01 | 重庆市银钢通科技有限公司 | A kind of chatter mark detection method and device |
| CN109074707A (en) * | 2016-04-20 | 2018-12-21 | 美高森美半导体(美国)股份有限公司 | glass breakage detection system |
| CN109187743A (en) * | 2018-08-27 | 2019-01-11 | 深圳市刻锐智能科技有限公司 | Glass breaking detection method, glass breaking alarm and storage medium |
| CN110322886A (en) * | 2018-03-29 | 2019-10-11 | 北京字节跳动网络技术有限公司 | A kind of audio-frequency fingerprint extracting method and device |
| CN111292767A (en) * | 2020-02-10 | 2020-06-16 | 厦门快商通科技股份有限公司 | Audio event detection method and device and equipment |
| CN116662729A (en) * | 2023-08-02 | 2023-08-29 | 山东鲁玻玻璃科技有限公司 | Low borosilicate glass feeding control data intelligent monitoring method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5441099A (en) * | 1977-09-08 | 1979-03-31 | Matsushita Electric Ind Co Ltd | Glass plate breakage detector |
| CN102610228A (en) * | 2011-01-19 | 2012-07-25 | 上海弘视通信技术有限公司 | Audio exception event detection system and calibration method for the same |
-
2015
- 2015-05-12 CN CN201510237822.6A patent/CN104865313B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5441099A (en) * | 1977-09-08 | 1979-03-31 | Matsushita Electric Ind Co Ltd | Glass plate breakage detector |
| CN102610228A (en) * | 2011-01-19 | 2012-07-25 | 上海弘视通信技术有限公司 | Audio exception event detection system and calibration method for the same |
Non-Patent Citations (3)
| Title |
|---|
| 何佳若等: "声信号的可视化特征提取方法", 《数字信号处理》 * |
| 刘鹏: "基于声谱图的公共场所异常声音特征提取及识别研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 栾少文: "智能监控系统中公共场所异常声音检测的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109074707A (en) * | 2016-04-20 | 2018-12-21 | 美高森美半导体(美国)股份有限公司 | glass breakage detection system |
| CN108098586A (en) * | 2017-12-25 | 2018-06-01 | 重庆市银钢通科技有限公司 | A kind of chatter mark detection method and device |
| CN110322886A (en) * | 2018-03-29 | 2019-10-11 | 北京字节跳动网络技术有限公司 | A kind of audio-frequency fingerprint extracting method and device |
| CN109187743A (en) * | 2018-08-27 | 2019-01-11 | 深圳市刻锐智能科技有限公司 | Glass breaking detection method, glass breaking alarm and storage medium |
| CN109187743B (en) * | 2018-08-27 | 2021-04-13 | 深圳市刻锐智能科技有限公司 | Glass breakage detection method, glass breakage alarm and storage medium |
| CN111292767A (en) * | 2020-02-10 | 2020-06-16 | 厦门快商通科技股份有限公司 | Audio event detection method and device and equipment |
| CN111292767B (en) * | 2020-02-10 | 2023-02-14 | 厦门快商通科技股份有限公司 | Audio event detection method and device and equipment |
| CN116662729A (en) * | 2023-08-02 | 2023-08-29 | 山东鲁玻玻璃科技有限公司 | Low borosilicate glass feeding control data intelligent monitoring method |
| CN116662729B (en) * | 2023-08-02 | 2023-10-31 | 山东鲁玻玻璃科技有限公司 | Low borosilicate glass feeding control data intelligent monitoring method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104865313B (en) | 2017-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104865313A (en) | Method and device for detecting glass breaking based on sound spectrum stripes | |
| CN101551729B (en) | Infrared touch scanning method and control device, infrared touch system of the same | |
| MX347748B (en) | Apparatus configured to detect gas usage, method of providing same, and method of detecting gas usage. | |
| WO2011062769A3 (en) | Methods and apparatus for performing capacitive touch sensing and proximity detection | |
| EA033063B1 (en) | Electrical event detection device and method of detecting and classifying electrical power usage | |
| WO2009102977A3 (en) | Systems and method for computer equipment management | |
| EP2570897A3 (en) | Edge grip detection method of a touch panel and a device using the same | |
| CN1996221A (en) | Display device with touch sensitive surface | |
| CN101762777A (en) | Local discharging on-line detecting and positioning system for electric transformer | |
| CN102288882A (en) | Partial discharge monitoring system and method for multichannel electrical equipment | |
| CN105630227A (en) | Sensitivity adjustment method and system for touch screen | |
| CN103123549B (en) | Touch-screen system and the synchronous method detected thereof | |
| WO2013013896A3 (en) | Locator | |
| WO2012007909A3 (en) | Analyzing ultrasonic signals using a dynamic signal processing window for an early detection of scaling in water processing equipment | |
| CN106846742B (en) | Fault trend method for early warning and system based on Hausdorff distance | |
| CN104698375A (en) | SF6 breaker electrical endurance monitor system based on weak signal processing | |
| CN105628710A (en) | Material detection system and detection method thereof | |
| WO2011031589A3 (en) | Methods and arrangement for detecting a wafer-released event within a plasma processing chamber | |
| CN103713793A (en) | Locating method for self-capacitance type touch screen | |
| CN102706882A (en) | Visual inspection method for SIM (Subscriber Identity Module) | |
| CN102236475B (en) | Touch positioning method and device, touch screen and display | |
| CN202904536U (en) | Laser touch screen | |
| GB201004376D0 (en) | detecting at least one community in a network | |
| CN202725680U (en) | Detection device for punch holes of lead frame | |
| CN106156703A (en) | Fingerprint acquisition apparatus and method for sensing thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |